Archived site. This is an independent restoration of material previously published on selsickis.com. It is not affiliated with, endorsed by, or operated by any organisation named on these pages, and the content is historical — it is not current information. About this archive

Selsick Investment Solutions

Innovative thinking…Systematic implementation

Registered Investment Advisor · Manhattan Beach, California

Coastal horizon strip used as the site's earlier page header

Solutions

The Enduring Portfolio Solution

Building the Enduring Portfolio – Strategic plus Tactical

Based on our research spanning more than 40 years of historical data, we understand how different assets perform during varying stages of the economic cycle. The Enduring Portfolio Solution starts with a carefully chosen strategic mix of asset classes, and then dynamically adjusts asset class exposure for each stage of the economic cycle. The model would have provided close to a 50% improvement in risk adjusted returns over the 60/40 benchmark.

1. Start with a sensible strategic allocation

The goal of the strategic asset allocation process is to achieve a risk balanced portfolio that is not concentrated toward any one driver of returns. It is a strategic allocation to assets that you think will meet your objectives over the long term. Different classes of assets will have varied performance in different economic environments, and since it is not possible to know what the future holds ten years from today, it is prudent to diversify your assets so that you are not beholden to any one future scenario.

Our process starts by selecting a strategic asset allocation mix using stocks, bonds, gold, commodities and real estate. It begins with a careful assessment of how each asset class performs through various economic environments. Our proprietary economic cycle classification system enables this. We also give consideration to how the current macro environment differs from the past 40 years on major issues such as demographics, the long term debt cycle, and the global investment opportunity set. You can find a more extensive explanation of our thinking on the strategic allocations for each asset class here.

2. Tactically adjust as the cycle evolves

Economic cycles generally last 5-7 years, and while it may not be possible to know the future 10 years from now, we can determine with some accuracy, the stage of the cycle we are currently in. This is again based on our proprietary methodology to measure and classify the stages of the cycle in real time. Knowing the current stage of the cycle, and how different assets perform during each stage of the cycle, drives the process of tactically deviating from the strategic allocations.

One additional value-added feature of our tactical adjustment process is a concept we call “true” bond exposure. In this article we explain the concept of true bond exposure and the mechanics of its calculation and implementation. The basic idea is that owning stocks includes a short exposure to bonds to the extent that the enterprise is carrying debt on its balance sheet. This is offset against your nominal bond exposure to give you a true net bond exposure. By targeting a true bond exposure, we were able to achieve better risk adjusted returns historically, and can more accurately tailor our bond exposure for different stages of the economic cycle.

Together, our process significantly improves the risk- adjusted returns over the 60/40 benchmark. It is a rules-based solution, time tested over many economic cycles, and implemented with no discretionary overrides.

Portfolio Construction Process

Combining the strategic and tactical elements as follows:

  • Our allocation starts by assigning to real estate first, followed by, gold and commodities, with the balance divided between stocks and bonds.
  • Real estate receives a substantial 17.5% permanent allocation and the balance is distributed among the other asset classes based on the stage of the economic cycle and our target bond exposure.
  • Next, gold and commodities receive an allocation of 7.5% and 5% respectively. However, this is reduced during certain stages of the economic cycle, with the reduced portion being allocated to stocks and bonds.
  • After real estate, commodities and gold, the balance (70.0% – 82.5%) is divided between stocks and bonds based on our target bond exposure formula (our goal is to keep our bond exposure to a necessary minimum for safety).
  • Lastly we determine the allocation breakdown within asset classes, for example, government versus corporate bonds, US equities versus foreign equities and so on.

Applying our formula would have resulted in the following exposures for each asset class for the past 40 years:

Performance

Using this portfolio construction formula, the following table shows performance data for the period, starting in 1973, the year from which data was available for all of the asset classes, through 2014.

Returns shown are average annual returns, not compound returns (compound returns can give a distorted picture depending on the start and end dates chosen so we avoid them). Our methodology would have significantly outperformed the 60/40 benchmark for the past 40 years on a risk adjusted basis. The model represents a 48% improvement in the risk adjusted returns over the 60/40 benchmark.

It also takes into account the realities of today, looking forward, to design a strategic allocation mix that makes sense for the next 30 years. Part of this entailed keeping our bond exposure to a minimum in the back-test, a period which included a 30-year bull market in bonds, meaning our model did well in spite of the fact that we minimized our bond exposure.

In summary, it is a strategic allocation that changes tactically over time based on our determination of which stage of the cycle we are in and our changing bond exposure. The model is dynamic wherein rebalancing happens immediately if the cycle changes, or else at least quarterly.

Implementation

The above results, based on monthly data are useful for testing the validity of an idea over long time-frames. The next step was to test it on daily data with real tradeable instruments. ETF’s for all the asset classes we use are available starting in 2004, so we test how the model would have performed for the period 2004 – 2014 using actual data. This chart shows the Enduring Portfolio measured against the benchmark 60/40.

The results confirm and exceed the performance of the monthly data model.

See the Enduring Portfolio Solution Factsheet for more details.